MR-guided laser induced thermal therapy (MRgLITT) for treatment of cancerous lesions in the brain presents a minimally invasive alternative to conventional surgery, gaining use worldwide with multiple on-going clinical trials currently. These therapies incorporate real-time MR temperature imaging (MRTI) to provide feedback which makes these procedures safe and feasible. However, as these therapies translate into clinical studies, there are two deficiencies worth addressing. Firstly, laser heating in the presence of tissue interfaces and convective heat transfer (e.g., ventricles, vessels and tissue perfusion) may render planning of treatments by assuming symmetric ellipsoidal lesion generation difficult, particularly when multiple applicators with arbitrary placement are used. Second, MRTI information may become corrupt in some regions (e.g., from temperature dependent signal losses, blood, or susceptibility effects at interfaces) and the temperature information may be lost or become extremely uncertain, making decisions about treatment outcome difficult. Therefore, in order to enhance the safety, efficacy and conformality of treatment delivery, there exists a critical need for prospective 3D treatment planning of MRgLITT procedures as well as more robust real-time monitoring to complement MRTI. The primary goal of the proposed research is to develop and validate algorithms to provide both prospective 3D treatment planning and real-time model driven treatment monitoring for MRgLITT procedures in the brain. Such a system will incorporate the use of 3D MRI acquisitions currently used for neurosurgical and stereotactic treatment planning and allow the user to interactively navigate the tissue and simulate the effects of various applicator trajectories and laser exposures, update these plans once the laser is positioned, as well as provide real-time 3D estimation of damage based on the acquired multi-slice MR temperature imaging input. Completion of this project will provide not only the tools outlined above, but a framework for our long term goal of applying these models for computer optimized inverse treatment planning as well as real-time modeling for adaptive control of therapy. This project is both innovative and timely in that no technology presently exists for prospective 3D planning of this emerging therapeutic option, much less, integration of such an algorithm with data driven model prediction assistance for monitoring and treatment assessment. This research is high risk in that the complexity of the computational problems to be addressed in near real-time present both physical and theoretical hurdles to overcome, but our previous experience in this area indicate it is feasible. PUBLIC HEALTH RELEVANCE: MR-guided laser induced thermal therapy (MRgLITT) for the treatment of cancerous lesions in the brain is a minimally invasive alternative to conventional surgery and is now available as a commercialized product with many clinical trials in operation or beginning. The primary goal of the proposed research is to develop, investigate and validate 3D prospective treatment planning and real-time MRTI driven model prediction assisted monitoring for MRgLITT procedures in the brain in order to enhance the efficacy and safety of these procedures. This project is both innovative and timely in that no technology presently exists to support prospective 3D treatment planning of this emerging therapeutic option nor data driven real-time modeling for prediction of temperature and tissue damage.